Discrete Adaptive Fuzzy Gradient-Based Control for Robotic Manipulators

This paper is devoted to the discrete-time position control of electrically driven robot manipulators. A novel discrete-time decentralized adaptive fuzzy controller is designed for each motor of the robot. The adaptive fuzzy controller is in the form of Takagi-Sugeno-Kang (TSK) adapted by a gradient descent algorithm. In estimation of uncertainty, the problem with gradient descent algorithm is that the training data is not available. To solve this problem, this paper finds a relationship between the tracking error and the estimation error. As an advantage, no data of the function of uncertainty is needed to approximate the uncertainty. The majority of guaranteed-stability control algorithms for robot manipulators require all feedbacks from the state variables. The design is easy to implement since it has the decentralized structure and uses only joint position, yet guarantees stability. The control design is a voltage-based control which differs from the torque-based control approaches. The novelty is that the proposed control algorithm eliminates the need for the velocity measurement and the motor current measurements.